current cold1 run sequence produces zero for ocean imports on first coupling timestep

[DeniseWorthen]

Describe the bug
The currrent cold1 run sequence ("coldstart") produces 0.0 for ocean import fields on first coupling timestep when restarting from coldstart

To Reproduce
All current implementations of ufs-s2s-model in regression tests and prototype workflows using the existing coldstart run sequence are impacted.

Steps to reproduce the behavior:

Use dbug >1 in mom_cap to turn on state_diagnose function in MOM6 and run any reststart-from-coldstart regression test. Examine the PET logs from Ocn at first ModelAdvance.

Expected behavior
Ocean imports at first coupling timestep should not be zero

Additional context
See attached presentation for further documentation of issue and proposed fixes.

remove_2step_coldstart.pdf

Also documented here: #86

[DeniseWorthen]

@junwang-noaa Should I put the fix for the existing 2-step cold start (rearrange the cmeps cold run sequence) in the s2s-cice6 commit? Otherwise, an entire new baseline will need to be created if we first want to fix the 2-step before making it optional.

[junwang-noaa]

That would be good.

…

On Thu, Sep 3, 2020 at 9:47 AM Denise Worthen ***@***.***> wrote: @junwang-noaa <https://github.com/junwang-noaa> Should I put the fix for the existing 2-step cold start (rearrange the cmeps cold run sequence) in the s2s-cice6 commit? Otherwise, an entire new baseline will need to be created if we first want to fix the 2-step before making it optional. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#167 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AI7D6TLPVGEVI4FL2XQJRQLSD6M47ANCNFSM4QIB4KMA> .

[junwang-noaa]

Is the s2s-cice6 commit ready to commit after this change? On Thu, Sep 3, 2020 at 9:54 AM Jun Wang - NOAA Federal <jun.wang@noaa.gov> wrote:

…

That would be good. On Thu, Sep 3, 2020 at 9:47 AM Denise Worthen ***@***.***> wrote: > @junwang-noaa <https://github.com/junwang-noaa> Should I put the fix for > the existing 2-step cold start (rearrange the cmeps cold run sequence) in > the s2s-cice6 commit? Otherwise, an entire new baseline will need to be > created if we first want to fix the 2-step before making it optional. > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub > <#167 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AI7D6TLPVGEVI4FL2XQJRQLSD6M47ANCNFSM4QIB4KMA> > . >

[DeniseWorthen]

I was ready to start creating baselines. I am just waiting for the go-ahead that I should start. I think Jessica is part-way through her baseline creation.

I also need the NEMS commit for the CICE6 commit. Since Jessica is relying on the reference to the current MOM6 directories in component_MOM6.mk for her commit, it is probably best to go w3, then cice6.

[junwang-noaa]

Yes, that is what we discussed at yesterday's meeting. I haven't seen any update in #171, I am checking with Jessica.

…

On Thu, Sep 3, 2020 at 10:00 AM Denise Worthen ***@***.***> wrote: I was ready to start creating baselines. I am just waiting for the go-ahead that I should start. I think Jessica is part-way through her baseline creation. I also need the NEMS commit for the CICE6 commit. Since Jessica is relying on the reference to the current MOM6 directories in component_MOM6.mk for her commit, it is probably best to go w3, then cice6. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#167 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AI7D6TOPAWOVKADNLE2XPGTSD6OO3ANCNFSM4QIB4KMA> .